FileMaker and JSON: keeping data and changes readable
By Oliver Schmalzl · Written: · Archive discovery date:
17 October 2024 · Archive date
When an application grows, its screen is not the only thing that matters. Data structures and changes should stay understandable too. Otherwise you later know that something is different, but not what has changed.
A small invented book file is enough to show the connection: a title has a text value, a volume number has a number value, and a list has several entries. A clearly described structure stops everyone involved from reading something different into the same fields.
Data needs a readable form
JSON describes objects and lists with explicitly named values. That helps when handing data between tools. For an exercise, field names, expected data types and how to handle empty values should be fixed beforehand.
If you use an extra shorthand notation, you have to check that converting it produces the same structure. Fewer characters are only an advantage if the meaning is kept.
For comparing structures, FileMaker offers an XML export. Claris explains "Save a Copy as XML" and points out that comparisons need the same export version. For a public demo, a small self-made file is enough; company data or scripts are not needed.
Interfaces do not change who is responsible
An AI connection via MCP can make tools and data access available. The MCP architecture describes the roles of host application, client and server. Which functions a specific FileMaker connection actually offers is a separate product question.
To start with, a read-only query of the invented book list would be easy to follow. After that you can check whether fields are understood correctly and changes are described cleanly. The readable structure stays just as important as a convincingly worded answer.
Original finds
These links were the starting point for this post. The assessment in the text is my own; the linked posts are not my own tests.
- Opening up FileMaker files with fmptools
- DDRparser: following FileMaker structures and scripts
- DDRparser: following FileMaker structures and scripts
- FileMaker performance as a learning topic
- Tracking FileMaker changes with XML and FMUpgradeTool
- Claude Code documentation as a starting point
- Moving AI scripts between a coding editor and FileMaker
- Moving AI scripts between a coding editor and FileMaker
- fmJAML: working with JSON more readably
- Preparing FileMaker solutions for agent work
- Claris Docs Skills for targeted help articles
- FileMaker scripts and version history
- Calling ChatGPT from FileMaker
- Talk on connecting a FileMaker solution to Claude via an MCP server.
- FM Analyzer analyses uploaded DDR XML files; do not upload your own company files.
- ZippyCue analyses DDR files locally on macOS and exports scripts, functions and structure information.
- Overview of AI development with MBS: local models, CoreML, Windows ML, OCR, Vision, Python and CURL.
- Podcast about ProofKit MCP and AI-assisted FileMaker development.
- The author uses ProofKit to demonstrate an AI-assisted FileMaker app for farming workflows.
- FMS Detective with log analysis, DDR import, an AI assistant and additional iOS management.
About the date: This post was written on 11 October 2026. The archive date is the date of the earliest related find in the collection, not of a publication at that time. Technical assessments and sources added later refer to the time of writing.